Characterization of Bamboo Culm for the Production of Laminated Beams

Transcription

1 Characterization of Bamboo Culm for the Production of Laminated Beams Juan Simón Obando 1,a and Khosrow Ghavami 2,b 1 Pontificia Universidade Católica do Rio de Janeiro, Brazil 2 Pontificia Universidade Católica do Rio de Janeiro, Brazil a jobando18@hotmail.com, b ghavami@puc-rio.br Keywords: Bamboo, Laminate Glued Bamboo (LGB), laminated beam, bamboo characterization, Non-Conventional materials. Abstract. Laminated bamboo was created to standardize the raw material in order to increase its strength, control its shape and develop sustainable and innovative structural elements. Bamboo is a Functionally Graded Material (FGM) due to the progressive distribution of the fibers across its wall thickness. This research presents the results of an experimental investigation series in which bamboo culm, of Dendrocalamus giganteus, was divided into 6 segments of analysis. Three divisions along its length, bottom, middle and top, and then two divisions across its wall thickness, inner and outer. In the first series, the specimens of each segment were tested separately to establish their tensile modulus of elasticity Et. Six types of bamboo uniaxial-laminated beam specimens of 2.5 cm width, 5cm height and 50 cm length were assembled with layers from each particular segment of bamboo culm, using resin of mamona adhesive. Four point bending tests were conducted on beam specimens to establish the bending modulus of elasticity Eb. Experimental values of both specimen groups were compared to those of theoretical values, applying solid mechanics theory. The results provide information to improve the segment arrangement of bamboolaminated beams upon subjection to bending loads. Based on the results, it is also possible to introduce equivalent values for the analysis of the mechanical properties of the beams, using solid mechanics theory. Introduction Scientific research on bamboo, oriented towards its physical, mechanical and microstructural properties study, has fostered its large-scale use as an engineering material. Based on its performance, there is a growing interest to make it available in shapes more suitable to current structural applications. Previous studies on this topic [1][2][3][4] assessed bamboo laminated boards made of strips taken indiscriminately from along the culm. However, bamboo is an anisotropic material as its properties vary in all directions. This condition restricts the analysis via traditional solid mechanics theory, and transmits bamboo heterogeneity to the laminate composite. Based on a pattern of property variation, it is possible to suggest divisions along the longitudinal axis initially, then across the wall in order to obtain less heterogeneous segments. Based on this observation it is possible to obtain groups of segments with similar properties and simultaneously heterogeneous among them. Lamination solves shape and some anisotropic issues of bamboo culm but also increases the cost, labor and equipment, which generate obstacles for local and decentralized production. Furthermore, variation of properties between different species requires considerable research for structural applications. This demands detailed investigation on local environments and restricts the use of worldwide information almost as benchmark only. Moreover, stable properties of those segments leads to maintain strip properties into laminated beam. Length and wall division arises as an alternative to produce bamboo-laminated beams, which could improve its bending properties by strips arrangement and their place of origin. This provides a deep characterization on local species Dendrocalamus giganteus.

2 Current experimental procedures Specific gravity and bending properties of bamboo vary with age and height location as well as cross the layer. They all increase from one-year-old bamboo to five-year-old bamboo until attains age for cutting [6]. Literature stated that outer layer had significantly higher SG and bending properties than the inner layer. This because, outer layer supports bamboo and strength wind loads. Bending strength had a strong correlation with SG [7] and so that with volume fraction [5]. Furthermore, height location determines its physical and mechanical properties [8] other studies about variation in mechanical properties of moso bamboo established an equation for predicting the tensile strength and modulus of elasticity from the position on the wall [9] similar with interpolation established by Ghavami [5]. To understand variations along the bamboo wall, Yu et all carried out testes which presented relative density ranging from 0,553 g/cm3 at internal edge to 1,006 g/cm3 at external one [10]. The mean longitudinal tensile MOE ranged from 8,987 to 27,397 GPa and mean longitudinal tensile strength ranged from to MPa, both from inner wall outwards. Layer and height position have a significant effect on all of those studied properties except for tensile strength. Relative density, tangential shrinkage, tensile MOE and tensile strength of bamboo increase greatly from inner layer outwards. Longitudinal shrinkage decreased greatly from the inner layer outwards, relative density, tangential shrinkage and tensile MOE at 1,3m were less than those are at 4,0m height [7]. Compression properties parallel to the longitudinal direction are significantly higher than perpendicular to the longitudinal direction therefore, it makes bamboo aligned longitudinally an optimal structural material for compression strengths [7]. Verma and Chariar carried out a layered laminate bamboo composite study, analyzing mechanical properties in segmented sectors along the culm and wall thickness [11]. It can be seen how stiffness and strength increase from lower internodes to top and a considerable difference between inner and outer regions. Middle regions in some cases presented higher stiffness than outer region, but strength always increase from center outwards. They found all laminas presented a bilinear stress-train curves and tensile strength and modulus of elasticity increasing from inner to outer region and from bottom to top of culms. Experimental results had good agreements between the estimated and predicted value for modulus of elasticity and tensile strength, which are satisfactory agreement for initial design purposes [12]. Experimental methodology There were used two types of specimens, test specimens and beam specimens. Test specimens were thin individual layers 200 mm length to determine tensile modulus of elasticity Et and shear modulus G for each segment of analysis (BI, MI, TI, BO, MO and TO). Beams specimens were elements composited by layers exclusively from a segment of analysis to determine bending modulus of elasticity Eb. 3 culms were cut in approximately 40 cm above ground level. After curing and drying, bamboo culm was divided in three parts along its length (Bottom, Middle and Top) approximately 2.5 to 3.5 m each part. Then transversal sections were split radially obtaining 5 to 8 strips 3 to 4 cm width and cut longitudinally 120 cm length. Those strips were processed in order to remove knots and irregular surfaces. All the segments were labeled: front view cross section view Bottom-Outer (BO), Middle-Outer (MO), Top-Outer (TO), Bottom-Inner (BI), Middle-Inner (MI) and Top-Inner (TI). Their longitudinal labeling and cross section location are shown in Figure 15. TO MO BO TI MI Figure 1 Segments of analysis depending on the longitudinal and cross section location. BI

3 3 mm 10mm Test specimens. There were obtained 18 test specimens for the modulus of elasticity and 18 more for shear modulus tests (3 samples for each segment of analysis), those were obtained from bamboo culms by a radial-symmetrical cut in the cross section along the culm. This provided strips with an accurate isosceles trapezium cross-shape. As a first step, planner thicknesser was used to remove both curve surfaces in order to plane strips. After that, the planner saw polished layers edges, shaping a square form for the transversal section. The laminated vertically was used in this research, applying loads on the stack direction. Inner and outer segments compose the final strip. Outer layer was obtained by sanding inner surface layer outwards until get 3mm thickness and the same process was carried out for the outer wall. Final thickness of layers was approximately 3mm Figure 2. Those layer were raw material for samples and laminas to assemble beams. Figure 2 Splitting longitudinally bamboo culm to obtain strips, then they are process to reduce and flatten strips into test specimens Tensile modulus of elasticity was determined for each segment of analysis, although according to standard ISO/DIS 22157, for simple tensile test parallel to fibers, this should only be done with bottom culms. Deformations were obtained by using one clip-gage placed at the center of the test specimens, and three of them using both clip and strain-gages to calibrate readings. Three test specimens of 200mm long, 100mm width and 3mm thick were obtained for each segment of analysis. Ghavami suggested straight shape test specimens, as in almost all cases shapes are narrow in the middle (known as bone shape) to force failing in this area. However, those assumptions generate stress concentrations in fiber-matrix composites that could alter the results. Samples were identified as UST (unnotched simple traction). Sample shape and dimensions are shown below. 200 mm Figure 3 Sample scheme for modulus of elasticity determination tests. Beam specimens - Laminated Glued Bamboo (LGB). There were used 18 beam specimens, 3 by each segment of analysis. They were assembled stacking randomly 10 to 14 layers of 550 mm approximately and effective span of 500 mm. Layered bamboo arrangements glued with any kind of adhesives have been named with many terms, depending on adhesives and the types of wood arranging them. Layers obtained in previous steps were arranged in sets with similar conditions, before being stacked using mamona resin as adhesive laminator. Strips were divided into groups of the same analysis segments, as seen on Figure 4 (a). First layers were Figure 4 (a) (b) (c)

4 arranged to spread resin uniformly, then second layers were placed aligned to the first ones, and this process was until having a stack of 8 to 12 pieces, as can be seen on Figure 4 (b). As the chemical reaction involved is exothermic, the drying process does not need heating. However, a pressure ranging between 15 to 20 kpa was applied during 24 hours after applying the adhesive. Blocks were polished in order to remove resin excesses and to shape longitudinal and cross-sections. Final beams ready to test are shown on Figure 4 (c). Beams were instrumented with upper and lower strain gages in the midpoint of the beam to measure strains. A LVDT sensor was located in the middle to read vertical displacement and loads applied by the load applicator. All sensors were connected to acquisition system, which recorded Figure 5 Set up of 4 point bending test values of each one during the test period. The Figure 24 shows the test arrangement for 4 points bend test, which is set up in this way to generate pure bending moment at the middle-span. Tests were carried out until failure although strain gages and LVDT in almost all cases were exceeded. However, information recorded was enough to determine the modulus of elasticity and shear modulus for the LBL casted.4 Different methodologies were used to establish bending modulus of elasticity. In the first place, modulus was established based on elastic curve equations. Then a methodology based on upper and lower strains on the beams by using strain gages. Finally, the ASTM provides equations to calculate directly the apparent modulus of elasticity and shear corrected modulus [13] on static test for lumber in structural sizes. Results Table 1 contains values of modulus of elasticity for both specimens with standard deviation and their specific gravity. Those values were obtained through ANOVA analysis where data was processed in order to generate marginal average values, principally due to number of samples. Test Specimens Beam Specimens Et [Gpa] s % s of Et SG Eb [Gpa] s % s of Eb BI 7,4 1,3 18% 0,64 0,69 MI 16,2 6,7 41% 0,74 15,8 7,4 47% 0,77 TI 12,0 8,2 68% 0,85 7,7 1,2 16% 0,86 BO 7,8 3,3 42% 0,63 5,9 4,6 78% 0,69 MO 16,1 7,1 44% 0,86 12,8 3,0 23% 0,87 TO 18,3 11,8 64% 0,92 12,1 1,1 9% 0,92 Table 1 mechanical properties of test and beam specimens produce during this study Figure 6 compares marginal average values in GPa of Et and Eb from same segment of analysis.

5 Figure 6 Marginal average values for Et and Eb [GPa] for all segments of analysis ASTM [13] for specimens of timber, explains the types of failure in static bending, as it was presented in the second chapter. However, in terms of fracture surfaces, bamboo may be roughly divided into brash and fibrous. The term brash indicates abrupt failure and fibrous indicates a fracture showing splinters. Moisture content is an essential factor to determine the type of fracture. Table 2 shows types of Specimen Primary type of failure Secondary type of failure Surface fracture 1 BI simple tension horizontal shear fibrous 2 BI simple tension horizontal shear fibrous 3 BI splintering tension horizontal shear fibrous 4 MI splintering tension inner compression fibrous 5 MI splintering tension inner compression fibrous 6 MI splintering tension inner compression fibrous 7 TI simple tension inner compression brash 8 TI simple tension inner compression brash 9 TI brash tension inner compression brash 10 BO horizontal shear compression fibrous 11 BO horizontal shear compression fibrous 12 BO brash tension horizontal shear brash 13 MO horizontal shear compression brash 14 MO horizontal shear compression brash 15 MO horizontal shear compression brash 16 TO brash tension horizontal shear brash 17 TO brash tension inner compression brash 18 TO brash tension inner compression fibrous Table 2 Types of primary and secondary failures on the beams and surface fracture. failure of each laminated bamboo beam. Moreover, it presents an alternative secondary cause of failure, related to failures that preceded the initial fracture during the tests. Another classification in terms of surface fracture is included as well. The results presented a marked trend to fibrous surface fracture on the inner specimens and brash surface fracture for outer ones. Inner specimens failed due to tensions strengths considered simple, splintering and brash tension with a great presence of horizontal shear and inner compression strengths as secondary causes of failure. Outer specimens presented a predominant horizontal shear and brash tension as primary cause of failure, and compression with horizontal strengths as secondary cause for inner specimens. Conclusions Bamboo-Laminated Beams, of Dendrocalamus giganteus species, assembled with middle (both inner and outer sections) segments generate the higher resistance to bending loads. Top segments should be used as whole layer (both inner and outer), without any split. Bottom segments are suggested to compose elements subjected to lower bending loads or axial strengths. The Split results helpful to improve design accuracy or reduce heterogeneity. As previous studies in other species, shear modulus does not present a significant variation. Relation between Et and Eb, for bending calculations, due to less variation of segments properties. Accurate equivalences introduced using solid mechanics approach require a wide and segmented characterization of materials

6 Despite mamona s resin presented a suitable adhesive performance, fractures due to shear indicates that it is not a suitable adhesive to be used on lamination process. Further research is required to include factor such as ply orientation on lamination, adhesive used, layered combination, and moisture content, to perfect the approach presented and allow scaling-up to small scale commercial manufacture. References [1] Sulastiningsih, & Nurwati. (2009, July). Physical and mechanical properties of laminated bamboo board. Journal of Tropical Forest Science, pp [2] Mahdavi M, Clouston P, Arwade S. Development of Laminated Bamboo Lumber: a review of processing, performance and economical considerations. ASCE J Mater Civ Eng 2011;23(7): [3] Correal, J. F., Ramires, F., & Yamin, L. E. (2009). Experimental study of Glued Laminated Guadua as building material: Adhesive calibration. 8th World Bamboo Conference Volume 8, (pp ). Bangkok, Thailand. [4] Yu, H. Q., Jiang, Z. H., Hse, C. Y., & Shupe, T. F. (2008). Selected Physical And Mechanical Properties of Moso Bamboo (Phyllostachys Pubescens). Journal of Tropical Forest Science, [5] Ghavami, K., & Marinho, A. (2001). determinação das propiedades mecanicas dos bambus das especies> moso, matake, guadua angustifolia, guadua tagoara e dendrocalamus giganteus, para utilização na engenharia.. Rio de Janeiro: PUC-Rio Publicação BMNC-2 Bamboo 02/2002. [6] Ghavami, K. (2003). Bamboo: Functionally Graded Composite Material. Asian Journal of Civil Engineering (Building and Housing), 4:1-10. [7] Li, X. (2004). Physical, Chemical, and Mechanical Properties Of Bamboo And Its Utilization Potential For Fiberboard Manufacturing. Beijing: The School Of Renewable Natural Resources. [8] Lee, A., Bai, X., & Peralta, P. (1994). Selected physical and mechanical properties of giant timber bamboo grown in South California. Forest Products Journal, [9] Xian, X., Shen, D., & Ye, Y. (1995). Bamboo Material and Bamboo Fiber Composite Materials. Sience Press Beijing. [10] (Yu, Jiang, Hse, & Shupe, 2008) [11] Verma, C. S., & Chariar, V. M. (2012). Development of layered laminate bamboo composite a mechanical properties. Elsevier, [12] Verma, C. S., & Chariar, V. M. (2013). Stiffness ans strength analysis of four layered bamboo composite at mircoscopic scale. Elsevier, [13] ASTM. (2014). Standard Test Methods for Small Clear Specimen of Timber D143. West Conshohocken, PA: ASTM international.